Guanine nucleotide and Mg2+ ion regulation of [125I-Tyr10]monoiodoglucagon ([125I]MIG) binding to liver plasma membranes from chicken, rat, and rabbit was studied. It was found that [125I]MIG binding to chicken liver membranes was increased by the addition of Mg2+ ion, while binding to rat and rabbit liver membranes was unaffected. In the chicken liver membranes, the Mg2+ ion induced high affinity binding which was sensitive to guanine nucleotides, while the low affinity binding in the absence of Mg2+ ion was not. Maximal effects of Mg2+ ion were observed at 1 mM. Glucagon binding to rat liver membrane receptors was GTP sensitive regardless of whether Mg2+ ion was added. Glucagon binding to rabbit liver membranes was insensitive to both Mg2+ ions and GTP. This lack of GTP effect was not due to degradation of GTP; no effect of the nonhydrolyzable analog guanyl-5'-yl-imidodiphosphate was observable. Glucagon stimulation of rabbit liver adenylyl cyclase, however, was dependent on GTP, as was the case with all of the other liver adenylyl cyclases studied here. The Kact of GTP for the rabbit liver system was very similar to that for rat liver membranes. The glucagon receptor was covalently labeled with [125I]MIG using p-hydroxysuccinimidyl azidobenzoate and analyzed by sodium dodecyl sulfate-gel electrophoresis. In all cases, a major labeled band at 63,000 daltons was observed. The levels of glucagon receptor and stimulatory (Ns) and inhibitory (Ni) regulatory proteins of adenylyl cyclase were measured. The highest levels of glucagon receptor were measured in rat liver membranes, while the levels in chicken and rabbit membranes were 30-40% lower. Rabbit liver membrane had the highest levels of Ns, while rat liver membranes had 2-fold lower and chick liver membrane 4-fold lower levels than rabbit liver membranes. The levels of Ni was similar in the three systems. Thus, the ratio of Ns to glucagon receptor was highest in the rabbit. In the rat, this ratio was 3-fold lower than that in the rabbit. In the chicken membranes, the ratio was about 60% of that in the rat. These data suggest that the observed differences in effects of GTP on hormone binding can be explained by alterations in the ratio of the receptor and Ns proteins among the various species.